Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with broad activation language, but it does not contain hidden execution, credential handling, persistence, or destructive behavior.
Installers should be aware that the skill may activate on broad words like self, improving, or proactive. It is otherwise a low-risk workflow helper, but narrower trigger wording would make it less likely to appear in unrelated conversations.
The manifest description says to use the skill whenever a user asks for broad concepts like work-productivity, self, improving, proactive, or practical support, which is ambiguous enough to match many benign requests. In an agentic system, such overbroad invocation guidance increases the chance that this skill is selected inappropriately and influences responses outside its intended scope.
The trigger list includes extremely generic terms such as "self," "improving," and "proactive," which are common in ordinary prompts unrelated to this skill. That makes accidental invocation likely and can cause the agent to apply the wrong workflow, leading to scope hijacking, irrelevant actions, or unintended automation in unrelated contexts.
The trigger list includes very broad everyday words such as 'self', 'improving', and 'proactive', which can match many unrelated user requests and cause accidental invocation. In an agent system, over-broad activation can route benign tasks into an unintended workflow, producing irrelevant actions, confusion, or unnecessary automation that the user did not ask for.
The trigger sentences and keywords are broad enough to match many ordinary user requests containing generic terms like 'self', 'improving', 'proactive', or 'practical workflow'. This can cause the skill to activate outside its intended scope, leading to inappropriate routing, unintended disclosure of context into the skill, or interference with more suitable skills.
The trigger phrases are extremely generic and can match ordinary user requests for help, causing the skill to activate outside its intended scope. In an agent ecosystem, overly broad activation increases the chance that this skill intercepts unrelated tasks and injects self-improving or proactive workflow behavior where it was not explicitly requested, which can lead to misrouting, unexpected actions, or instruction precedence issues.
The trigger phrases are extremely broad and include generic terms like 'self', 'improving', and 'proactive', which can cause the skill to activate in many unrelated contexts. In an agent ecosystem, overbroad activation can route user requests into unintended workflows, causing prompt/skill hijacking, incorrect task execution, or accidental handling of sensitive requests by the wrong skill.
The skill description defines usage conditions very broadly, covering many generic requests for workflows, checklists, analysis, or implementation support around loosely related concepts. This weak boundary definition increases the chance that the skill is selected outside its intended scope, which can degrade reliability and indirectly affect safety by bypassing more appropriate specialized skills.
The skill’s invocation description and default prompt are extremely broad, covering generic 'work-productivity' and 'self-improving/proactive agent' requests without narrow trigger conditions. Combined with allow_implicit_invocation=true, this can cause the platform to auto-invoke the skill in unrelated contexts, increasing the chance of unintended prompt injection exposure, overreach into user workflows, or execution of actions the user did not explicitly request.
The file declares English and Chinese variants for instructions and user guides, but does not state how the user's preferred language is selected or that language choice is optional. This can create a locale-policy concern if the skill defaults to one language without explicit user opt-in.
This file is presented as a Chinese README, but key user-facing content including the workflow description and trigger phrases is written in English only. That effectively imposes a language choice on the user without documenting opt-in or offering alternatives.
